Improve the management of game life cycle information using TAGs

[TheFapMasterOne]

Hi ,
what I've read in this post has made me want to suggest improvements for the management of game life cycle information using TAGs.


Original post of the stats here

With more than 12k games on this site, it becomes more and more difficult to follow in a meaningful way the progress and the evolution of some games, VN,.. and others... to better manage the favorites according to the frequency of games release, I suggest adding a Release Scale Tags!

it allows us to directly know how often the game is going to evolve and also to limit the useless discussion threads that only ask, "When is the next release?" ... because of the lack of information on the release, some games are wrongly rated.

Almost 3,5k abandoned games!



so, I propose this release scale that we could call it Release Frequency Scale: rated from 1 to 4: R1, R2, R3, R4

The R4 Tag needs more precision. because given the long release interval we are entitled to wonder if the game is dying or not! this is regardless of the promises and other declarations of the devs! The accuracy of the tag could be: abandoned, on hold, Pending, on Standby, need support, need ca$h, need assets, need resources,…

R1: Release every 1 to 3 months

R2: Release every 3 to 6 months

R3: Release every 6 to 12 months

R4: Release that takes more than a year


The point here is to bring precision and not confusion

 

[redknight00]

Not gonna happen, a dev's release schedule is their own business, we are not here to police that.

Plus trying to assign value to update frequency is useless and gay, shit happens a dev might update monthly but have troubles irl or with a particularly complex section and go for a 6-month circle once. Not to mention the actual content, why should we promote games with a monthly 15m update versus a 6h update once per year?

 

[Count Morado]

Hi ,
what I've read in this post has made me want to suggest improvements for the management of game life cycle information using TAGs.

View attachment 2392662
Original post of the stats here

With more than 12k games on this site, it becomes more and more difficult to follow in a meaningful way the progress and the evolution of some games, VN,.. and others... to better manage the favorites according to the frequency of games release, I suggest adding a Release Scale Tags!

it allows us to directly know how often the game is going to evolve and also to limit the useless discussion threads that only ask, "When is the next release?" ... because of the lack of information on the release, some games are wrongly rated.

Almost 35k abandoned games!

View attachment 2392622

so, I propose this release scale that we could call it Release Frequency Scale: rated from 1 to 4: R1, R2, R3, R4

The R4 Tag needs more precision. because given the long release interval we are entitled to wonder if the game is dying or not! this is regardless of the promises and other declarations of the devs! The accuracy of the tag could be: abandoned, on hold, Pending, on Standby, need support, need ca$h, need assets, need resources,…

R1: Release every 1 to 3 months

R2: Release every 3 to 6 months

R3: Release every 6 to 12 months

R4: Release that takes more than a year


The point here is to bring precision and not confusion

You're using my table - which was posted in direct opposition to a similar post of entitlement as yours that you present here? Wow. Just. Wow. My data went right over your head and you sir.

redknight00's post is spot on.

Also, it's not 35k abandoned games ... it's fewer than 3500.

 

[Hagatagar]

And who has to do that?
Who has to check if those release frequencies are accurate or are being met?
Who has to check every dev for their release schedule even if that dev might bail out before a next update?
Mods, Uploaders, Members?
Not to mention that many devs might not be able to keep their planed schedules for a variety of reasons.

This is not a tenable system, because it requires much work over something one has no control over and that probably has to be checked again with every update anyway.

 

[anne O'nymous]

And who has to do that?
Who has to check if those release frequencies are accurate or are being met?
Who has to check every dev for their release schedule even if that dev might bail out before a next update?

I propose to add a rule to the forum, one come back from the golden age of USENET: Just Do It !

Anyone who propose a change implying constant/regular/significant works from uploaders/moderators/members, volunteer to do it all by himself for at least the first three months.


In the current case, TheFapMasterOne thanks for the time you'll pass browsing the ~5,000 patreon/whatever pages corresponding to a WIP game, in order to compute the average delay between each updates. And, please, don't forget to weight the results accordingly to the average amount of content, this to reflect the spirit behind the last update of the rules.
Oh, and before I forgot, don't forget to take a subscription to the nearest psychic. You'll need it to know what will be the average delay between each updates, and their amount of content, for the hundredth games that will be started during those three months.

 

[TheFapMasterOne]

The point of this proposal is of course not to pinch the developers or to put red labels on them.
It was just a simple reflection about how to improve the issue of ‘Ghost Content” here.

Also, the provided numbers are not based on exact statistics about abandoned games, VN, ... By using the "abandoned "filter, we can find 3,5k/12K abandoned content.

I'm aware that there could be other factors contributing to the presence of Ghost Content on the site, and I'm not claiming to have all the answers. But, I believe my proposal could help address part of the problem by improving the way we can follow the progress and the evolution of some games, VN,.. and others.

Regarding tags management, I think it would be possible to implement an automated system if F95 gives access to the latest release or update date attribute of the published content. With a bit of code, we can automatically match suggested tags to the published and existing content.
I understand that my proposal may not be the best solution. However, I am still convinced that some ideas should be explored to improve the quality of the site and its content also deal with ghost content

 

[TheFapMasterOne]

I propose to add a rule to the forum, one come back from the golden age of USENET: Just Do It !

Anyone who propose a change implying constant/regular/significant works from uploaders/moderators/members, volunteer to do it all by himself for at least the first three months.


In the current case, TheFapMasterOne thanks for the time you'll pass browsing the ~5,000 patreon/whatever pages corresponding to a WIP game, in order to compute the average delay between each updates. And, please, don't forget to weight the results accordingly to the average amount of content, this to reflect the spirit behind the last update of the rules.
Oh, and before I forgot, don't forget to take a subscription to the nearest psychic. You'll need it to know what will be the average delay between each updates, and their amount of content, for the hundredth games that will be started during those three months.

Code can do magic

 

[TheFapMasterOne]

You're using my table - which was posted in direct opposition to a similar post of entitlement as yours that you present here? Wow. Just. Wow. My data went right over your head and you sir.

redknight00's post is spot on.

Also, it's not 35k abandoned games ... it's fewer than 3500.

The last search filter gives 3.5k abandoned content !

 

[TheFapMasterOne]

Not gonna happen, a dev's release schedule is their own business, we are not here to police that.

Plus trying to assign value to update frequency is useless and gay, shit happens a dev might update monthly but have troubles irl or with a particularly complex section and go for a 6-month circle once. Not to mention the actual content, why should we promote games with a monthly 15m update versus a 6h update once per year?

And who has to do that?
Who has to check if those release frequencies are accurate or are being met?
Who has to check every dev for their release schedule even if that dev might bail out before a next update?
Mods, Uploaders, Members?
Not to mention that many devs might not be able to keep their planed schedules for a variety of reasons.

This is not a tenable system, because it requires much work over something one has no control over and that probably has to be checked again with every update anyway.

Code my fella code!

 

[TheFapMasterOne]

Not gonna happen, a dev's release schedule is their own business, we are not here to police that.

Plus trying to assign value to update frequency is useless and gay, shit happens a dev might update monthly but have troubles irl or with a particularly complex section and go for a 6-month circle once. Not to mention the actual content, why should we promote games with a monthly 15m update versus a 6h update once per year?

The point is not to police them but to improve and limit the existing ghost content here!
a game or VM, which releases 15 min of content every month is always better than nothing ! 15 min each month, it's 3h of content per year ! it's certainly more than the duration of an update that took a year to be published ! even the interest of this kind of long-release content disappear!
But I agree there are always exceptions!...

 

[Hagatagar]

Code my fella code!

And where does the code get its input from? Is it supposed to be a calculated average of the release frequencies posted here? But that would be far from the precise thing you want, it would only be a rough estimate, and would also only work if there are regular releases.

Or is the code supposed to pull it from any official announcement from the devs (in case it's one of the few devs with a reliable release schedule). But even then, the code needs to be tailored to each particular method the dev chooses to announce their release.

And regardless of that, your desired R4 precision would still require additional manual work.


For a feature request like this, you have to give more than just say:

Code can do magic

What is your idea how the information is collected and processed?

 

[anne O'nymous]

Code can do magic

I know, it's been more than 35 years that I code, near to 30 as my job. But tell me, how do your code will works ?

Here are the main constraints your code will have to works with:

• Devs release games all around the web, what mean that even with a bot crawling Patreon, SubscribeStar, itch.io, dlsite and steam, you wouldn't cover them all ;
• Counting on the "latest update" list is not possible, because not all updates are promoted ;
• Counting on the update of the thread opening is not possible, because the post with the leaked version need to be noticed by an uploader, or reported, for the said OP to be updated. Once I reported a link (for a final release) that was one and half year old ;
• Counting on a post leaking the game in the game thread is not possible, because the leaked version can will make between few minutes to many months before it reach this forum ;

So, first thing first, you'll have to mix all the data sources and compute the effective release date ; I'm nice, a +/- 1 day difference is considered insignificant.


But this don't change the main issues that your code will have to face once it have the release date:

• The tag is needed right from the starts.
  
  In top of doing magic, will your code also do divination ? It's the only way to know when the first update will be released.
  
  
• The tag will have to be updated each time the average delay between update will change.
  
  Ok, this a code can do it, if run by a staff member.
  
  
• If a dev have an accident, a grave illness, whatever, keeping him far from the keyboard for a really long time, the average delay between updates will lost all meaning.
  
  Of course, you can exclude this update from the computation, but it can only be done manually. Your code will not be able to browse through all the media used by the dev to know if it made an announce explaining why there were a delay, nor would he be able to process the text and understand that it is such announce. What mean that a human will still have to look at the result of the day, and investigate on all anomalies.
  
  
• This provide a totally useless information because the average delay between two updates need to be weighted by the average amount of content by updates.
  
  If it's probably possible to somehow simulate a play for Ren'Py, RPG Maker, Tyrano Builder, Twine, Unity, and few more, it isn't possible for Unreal, Game Maker, RAGS, and even less all the homemade engines.
  And, no, looking at the game size wouldn't give you the answer. Firstly because relying on a percentage of size increase is stupid ; the more a game have had updates, the lower will be the size difference with the previous update. If each update need 100MB, the first one will add 100% of content, the second 50%, and you'll be below 1% after the tenth one. And of course, and update can be heavy on CG and really low on effective content, or the opposite. Secondly because:
  • The size needed for 1 hour of content isn't the same for a text based game, a sprite based one, a game that rely on fully rendered CGs, and one on real time 3D.
    Therefore you'll still have to manually process all games when they are released, in order to know in what category they fall.
    
  • A "real game" (shooter, platformer, etc) can add levels that will reuse already available assets.
    A 20KB update can add tenth of those levels, while a 5MB will only release one level based on totally new assets. Therefore you'll have to process those games manually each time there's an update, because the changelog wouldn't give you this information, assumed that there will be a changelog.
    
  • There's games that switch from text based, or full render, to sprite based.
    Therefore you'll have to process all games manually each time there's an update, to verify if it's the case or not.
    
  • There's devs that compress their CGs, others that don't, and some that starts to compress after few updates.
    Therefore you'll have to manually process each game each time there's an update, to see if it's compressed or not.
    
  • There's games that add a new executable every now and then.
    By itself it's the easiest problem to solve.

But, don't be afraid but what I said, don't hesitate a single second...
Amaze us, come back in six months (three to write the code, three as proof of concept with 1,000 games) to show us that it's totally possible. I'm sure that if your data are accurate, and your game sample cover all the possibilities in terms of source of the game, game type, and game engine, the staff will consider your proposition.

By the way, learn how to use the "quote" button at the left of the "reply" one, in order to not flood a thread with X time the exact same kind of answer.

 

[TheFapMasterOne]

I know, it's been more than 35 years that I code, near to 30 as my job. But tell me, how do your code will works ?

Here are the main constraints your code will have to works with:

• Devs release games all around the web, what mean that even with a bot crawling Patreon, SubscribeStar, itch.io, dlsite and steam, you wouldn't cover them all ;
• Counting on the "latest update" list is not possible, because not all updates are promoted ;
• Counting on the update of the thread opening is not possible, because the post with the leaked version need to be noticed by an uploader, or reported, for the said OP to be updated. Once I reported a link (for a final release) that was one and half year old ;
• Counting on a post leaking the game in the game thread is not possible, because the leaked version can will make between few minutes to many months before it reach this forum ;

So, first thing first, you'll have to mix all the data sources and compute the effective release date ; I'm nice, a +/- 1 day difference is considered insignificant.


But this don't change the main issues that your code will have to face once it have the release date:

• The tag is needed right from the starts.
  
  In top of doing magic, will your code also do divination ? It's the only way to know when the first update will be released.
  
  
• The tag will have to be updated each time the average delay between update will change.
  
  Ok, this a code can do it, if run by a staff member.
  
  
• If a dev have an accident, a grave illness, whatever, keeping him far from the keyboard for a really long time, the average delay between updates will lost all meaning.
  
  Of course, you can exclude this update from the computation, but it can only be done manually. Your code will not be able to browse through all the media used by the dev to know if it made an announce explaining why there were a delay, nor would he be able to process the text and understand that it is such announce. What mean that a human will still have to look at the result of the day, and investigate on all anomalies.
  
  
• This provide a totally useless information because the average delay between two updates need to be weighted by the average amount of content by updates.
  
  If it's probably possible to somehow simulate a play for Ren'Py, RPG Maker, Tyrano Builder, Twine, Unity, and few more, it isn't possible for Unreal, Game Maker, RAGS, and even less all the homemade engines.
  And, no, looking at the game size wouldn't give you the answer. Firstly because relying on a percentage of size increase is stupid ; the more a game have had updates, the lower will be the size difference with the previous update. If each update need 100MB, the first one will add 100% of content, the second 50%, and you'll be below 1% after the tenth one. And of course, and update can be heavy on CG and really low on effective content, or the opposite. Secondly because:
  • The size needed for 1 hour of content isn't the same for a text based game, a sprite based one, a game that rely on fully rendered CGs, and one on real time 3D.
    Therefore you'll still have to manually process all games when they are released, in order to know in what category they fall.
  • A "real game" (shooter, platformer, etc) can add levels that will reuse already available assets.
    A 20KB update can add tenth of those levels, while a 5MB will only release one level based on totally new assets. Therefore you'll have to process those games manually each time there's an update, because the changelog wouldn't give you this information, assumed that there will be a changelog.
  • There's games that switch from text based, or full render, to sprite based.
    Therefore you'll have to process all games manually each time there's an update, to verify if it's the case or not.
  • There's devs that compress their CGs, others that don't, and some that starts to compress after few updates.
    Therefore you'll have to manually process each game each time there's an update, to see if it's compressed or not.
  • There's games that add a new executable every now and then.
    By itself it's the easiest problem to solve.

But, don't be afraid but what I said, don't hesitate a single second...
Amaze us, come back in six months (three to write the code, three as proof of concept with 1,000 games) to show us that it's totally possible. I'm sure that if your data are accurate, and your game sample cover all the possibilities in terms of source of the game, game type, and game engine, the staff will consider your proposition.

By the way, learn how to use the "quote" button at the left of the "reply" one, in order to not flood a thread with X time the exact same kind of answer.

Thanks for your detailed answer and for taking the time to elaborate on the possible solution as well as the constraints and difficulties that may arise.
I think you wanted to implement a much more relevant, precise, and dynamic solution than what I had specified and imagined at the beginning. It's obvious that if you want to have real-time data of the contents and publication on the different publishing sources, the task will be hard, maybe impossible.
Even if we had APIs access to all the publishing platforms and retrieve any data the processing and implementation of the solution would be very tedious. Especially if we go into details as you've well detailed.

My basic idea was simple: we only work with data and content available and published on the site. The data source will be F95 db only . The point is to organize the available content here not to create a versioning system of the multi-source contents of publication! here the task is tedious
The idea is to rely on 2 fields to assign these tags automatically to games.
- Release Date:
- Current Version
-Last Version
Then work on a data format as for example below:

Games	Release Date	Current Version	Last Version
Game1	12/02/2022	V 0.1	V 0.1
Game3	12/02/2023	V 0.5	V 0.5
Game4	09/6/2022	V 0.9	V 1.0

A js code of this nature allows the comparison of tags assignments!

let games = [
  { name: "Game1", releaseDate: "12/02/2022", currentVersion: "V 0.1", lastVersion: "V 0.1", tag: "" },
  { name: "Game2", releaseDate: "14/09/2021", currentVersion: "V 0.2", lastVersion: "V 0.2", tag: "" },
  { name: "Game3", releaseDate: "18/05/2019", currentVersion: "V 0.2", lastVersion: "V 0.5", tag: "" },
//ofc the games attributs here and other information are to be retrivied from f95 db!
....
];

const currentDate = new Date();

games.forEach(game => {
  const releaseDate = new Date(game.releaseDate);
  const monthsDiff = (currentDate.getFullYear() - releaseDate.getFullYear()) * 12 + (currentDate.getMonth() - releaseDate.getMonth());

  if (monthsDiff <= 3) {
    game.tag = "R1";
  } else if (monthsDiff <= 6) {
    game.tag = "R2";
  } else if (monthsDiff <= 12) {
    game.tag = "R3";
  } else {
    game.tag = "R4";
  }

  const lastVersion = parseFloat(game.lastVersion.substr(2));
  const currentVersion = parseFloat(game.currentVersion.substr(2));

  if (lastVersion > currentVersion) {
    ......

  }
});

console.log(games);

... the syntax is maybe not correct the Algo too!


But well let’s be realistic I think that, this story of tags, I admit might get people lost! and the solution is much more complex than expected.

I also wanted to add that I felt a slight antagonism from all the answers as well! the answers were systematic defensive and less argumentative! apart from a little bit yours! ..the 2nd answer I mean!..

Thank you for the exchange and your detailed comments

My bad for not correctly using the Quote

 

[anne O'nymous]

My basic idea was simple: we only work with data and content available and published on the site. The data source will be F95 db only .

What is the interest to deal with information that are well known as being inaccurate ?

 

[TheFapMasterOne]

The point is to better categorize the contents based on the tags (Facilitate content browsing with tags) and also to better use the search filters by using different accurate tags!...

 

[Hagatagar]

I also wanted to add that I felt a slight antagonism from all the answers as well! the answers were systematic defensive and less argumentative!

I guess you are referring at least partially to me, and that was not my intention. Sorry if it came across that way.
English is no my native language and I try to keep this kind of posts reather short.
My questions were genuine.

But this does not change the unreliability of those statistics if they are exclusively pulled from F95zone.
As anne O'nymous mentioned, if it's not a somewhat popular game, uploads are only made when someone requests or provides them (and was noticed). This will skew any data and this system wouldn't be about release frequencies but F95 upload frequencies.

 

[Meaning Less]

Currently the website doesn't even keep track of the "last version", it has been a planned feature since 2019.